Method for intelligently identifying drowsiness action of driver

The invention provides a method for intelligently identifying drowsiness actions of a driver. The method comprises the following steps: 1, acquiring a video stream in a driving process of the driver; 2, preprocessing the video stream to obtain gray level image information and optical stream image information of the video stream; and 3, taking the grayscale image information and the optical flow image information as input of a drowsiness action recognition model so as to obtain the drowsiness action of the driver. According to the invention, the optical flow image is also used as the input of the drowsiness motion detection model, and the motion information of the object is stored in the optical flow image, so that the detection accuracy is further improved. 本发明提供一种智能识别司机睡意动作的方法，包括以下步骤：步骤一：获取司机驾驶过程中的视频流；步骤二：对所述视频流进行预处理，得到该视频流的灰度图像信息和光流图像信息；步骤三：以所述灰度图像信息和光流图像信息作为睡意动作识别模型的输入，从而得到司机的睡意动作。本发明将光流图像也作为睡意动作检测模型的输入,由于光流图像中保存了物体的运动信息，从而使得检测的准确率进一步得到提高。